Pervasive Well-Being Technology
Total Page：16
File Type：pdf, Size：1020Kb
Pervasive Well-being Technology Pablo Paredes Electrical Engineering and Computer Sciences University of California at Berkeley Technical Report No. UCB/EECS-2016-37 http://www.eecs.berkeley.edu/Pubs/TechRpts/2016/EECS-2016-37.html May 1, 2016 Copyright © 2016, by the author(s). All rights reserved. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission. Acknowledgement I want to thank my advisor, Professor John F. Canny. I am grateful for picking me up among hundreds. Prof. Eric Brewer - You guided my passion with such bold vision. Thanks for making a putative son of TIER. Dr. Mary Czerwinski – I think we chose the most exciting and rewarding topic. Let's keep working at it. Prof. Greg Niemeyer - You are one of the kindest, most creative and interesting people I know. Looking forward to doing more together. Prof. Bjoern Hartmann - You became my “ad hoc” advisor and my main funding partner support for the latter part of my career. You are without any doubt on of the brightest minds in HCI. Prof. John Chuang - What a pleasure it was to work with you. I always admired your ability to bring together so many stakeholders to the wearables class. Pervasive Well-being Technology By Pablo Enrique Paredes Castro A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy In Computer Science in the Graduate Division of the University of California, Berkeley Committee in charge: Professor John F. Canny, Chair Professor Björn Hartmann Professor Greg Niemeyer Professor Mary Czerwinski Fall 2015 Pervasive Well-being Technology Copyright 2015 by Pablo Enrique Paredes Castro 1 Abstract Pervasive Well-being Technology by Pablo Enrique Paredes Castro Doctor of Philosophy in Computer Science University of California, Berkeley Professor John F. Canny, Chair Well-being is the characteristic of humans to feel well with themselves and their environment. A key driver of wellbeing is to have a healthy mind. Stress management, positive emotions, and empathic social interaction get us closer to this goal. The only sustainable way towards wellbeing is by maintaining a healthy lifestyle. In summary, technologies supporting this objective should measure and intervene "life" itself! The good news is that we are in a unique position to challenge the way technology can become a driver of wellbeing. Internet of things (IoT), Big Data, Affective Computing and Critical Engineering, are the key enablers. With pervasive data, interventions can evolve from being reactive to predictive. We could design long- term interventions that foster good habits, resilience, and personal growth. We encapsulate technologies in two parts: "Conceptualization and Sensor-less Sensing" and "Opportunistic Interventions." The former uses data streams to investigate and measure human traits and trends. The latter repurposes popular apps, devices, and environments into effective interventions. In summary, we aim at designing well-being technology that could survive the chasm of adoption and attrition. i Dedicated with love to my wife Claudia (Clau), my babies Manuela (Manu) and Diego Pablo, my parents Enrique and Cecilia, my brother Juan Carlos, my sister Maria Cecilia. You made this possible. Thanks forever! ii Contents ii List of Figures vi List of Tables x Introduction 1 Acknowledgements 3 Part 1: Conceptualization and Sensing: 5 1. Qualitative Insight Mining 7 1.1 INTRODUCTION ………………………………………………………... 8 1.2 RELATED WORK ………………………………………………………. 10 1.2.1 Keyword Search ………………………………………………… 10 1.2.2 Semantic Similarity ……………………………………………... 10 1.2.3 Qualitative Data Analysis Tools ………………………………. 11 1.3 METHODOLOGY ………………………………………………………. 11 1.3.1 LiveJournal Corpus …………………………………………….. 11 1.3.2 Data Preprocessing ……………………………………………. 11 1.3.3 word2vec ………………………………………………………... 12 1.3.4 Querying ………………………………………………………… 13 1.4 ITERATIVE CO-DESIGN PROCESS ………………………………… 15 1.4.1 Initial Concept Formation ……………………………………… 15 1.4.2 Early Idea Validation …………………………………………… 17 1.4.3 Low Fidelity (Lo-Fi) Prototype ………………………………… 17 1.4.4 Testing on more Data …………………………………………. 22 1.4.5 Mid Fidelity (Mi-Fi) Prototype …………………………………. 23 1.5 FUTURE WORK ……………………………………………………….. 26 1.6 CONCLUSION ………………………………………………………….. 27 2. Game & Narrative Design for Behavior Change 29 2.1 INTRODUCTION ………………………………………………………. 30 2.2 PREVIOUS WORK ……………………………………………………. 30 2.3 THE CHALLENGE: EFFICACY + EFFICIENCY ……………………. 31 2.4 DESIGN METHODOLOGY …………………………………………… 32 2.5 GAME PROTOTYPE EXAMPLES …………………………………… 33 2.6 NARRATIVE PROTOTYPE EXAMPLE ……………………………... 36 2.6.1 Mental Health Background ……………………………………. 36 2.6.2 Related Work …………………………………………………… 39 2.6.3 Prototype Movie: “A Rainy Day” ……………………………… 39 2.6.4 Uses of Therapeutic Machinima ……………………………… 42 2.6.5 Future Research ……………………………………………….. 42 iii 2.7 PRINCIPLES FOR CONCEPTUALIZATION ……………………….. 43 2.7.1 Understanding Disempowering Narratives ………………….. 43 2.7.2 Externalizing Problems ………………………………………... 43 2.7.3 Game Dynamics as Interventions ……………………………. 45 2.7.4 Narratives as Interventions ….………………………………… 45 2.8 CONCLUSION ..…..….………………………………………………… 45 3 Sensor-less Sensing 46 3.1 INTRODUCTION ……………………………………………………….. 47 3.2 BACKGROUND ………………………………………………………… 47 3.2.1 Arousal and Stress …………………………………………….. 47 3.2.2 Physiological Affect Sensing ………………………………….. 48 3.3 AROUSAL FROM COMPUTER PERIPHERALS …………………… 49 3.3.1 Mouse Movement ……………………………………………… 49 3.3.2 Mouse Pressure ………………………………………………... 52 3.3.3 Keyboard ………………………………………………………... 53 3.4 SENSING AFFECT WITH TEXT MINING …………………………… 57 3.5 MOVING FROM SENSING TO INTERVENTIONS ………………… 57 3.5.1 Usability Studies ……………………………………………….. 57 3.5.2 Stress and Emotion from Sensor and Contextual Data .……. 58 3.5.3 Integrated Model and Data Acquisition ……………………… 58 3.5.4 Causal Inference ………………………………………………. 58 3.6 ADAPTIVE INTERACTIONS AND INTERVENTIONS …………….. 59 3.6.1 Feedback and Interface Adaptation ………………………….. 59 3.6.2 Emotional Regulation and Psycho-education Interventions . 59 3.6.3 Foreground and Background Interventions …………………. 60 3.7 CONCLUSION …………………………………………………………. 60 Part 2: Opportunistic Interventions 61 4 Repurposing mobile and wearable systems 62 4.1 INTRODUCTION ………………………………………………………. 63 4.2 BACKGROUND AND RELATED WORK …………………………… 63 4.3 IMPLEMENTATION …………………………………………………… 63 4.3.1 Prototypes ……………………………………………………… 63 4.3.2 Common Use Scenario ……………………………………….. 66 4.4 STUDY DESIGN ………………………………………………………. 66 4.4.1 Objective Data …………………………………………………. 66 4.4.2 Subjective Data ………………………………………………… 67 4.5 EVALUATION ………………………………………………….. 67 4.5.1 Baseline ………………………………………………………… 67 4.5.2 Randomized Experiment ……………………………………… 68 4.6 RESULTS ………………………………………………………………. 68 iv 4.7 IMPLICATIONS FOR DESIGN ………………………………………. 71 4.7.1 Suite of Interventions and Context …………………………… 71 4.7.2 Design Improvements …………………………………………. 72 4.7.3 Long-term (Real-Life) Usability ………………………………. 73 4.8 CONCLUSION ...……………………………………………………….. 73 5 Harvesting web interventions 74 5.1 INTRODUCTION ……………………………………………………… 75 5.2 BACKGROUND WORK ………………………………………………. 76 5.2.1 Cognitive Behavioral Therapy (CBT) based Technology …. 76 5.2.2 Stress Technology Intervention R&D ……………………….. 76 5.3 STRESS MANAGEMENT SYSTEM DESIGN ……………………... 77 5.3.1 Micro-Intervention Authoring System ……………………..… 77 5.3.2 Intervention Recommender System ………………………… 80 5.3.3 Mobile App Interface …………………………………………. 83 5.4 STUDY DESIGN ……………………………………………………… 85 5.4.1 Phones and Screening Procedures ………………………… 85 5.4.2 Experiment Design …….……………………………………… 86 5.4.3 Gratuity and Incentive Policies ……………………………… 86 5.5 RESULTS ……………………………………………………………… 86 5.5.1 Descriptive Data ……………………………………………… 87 5.5.2 Qualitative Results …………………………………………… 91 5.5.3 Quantitative Results …………………………………………. 92 5.6 IMPLICATIONS FOR DESIGN ……………………………………… 94 5.6.1 Intervention Design …………………………………………… 94 5.6.2 Intervention Recommendation ………………………………. 95 5.6.3 Future Research ………………………………………………. 96 5.7 CONCLUSION ………………………………………………………… 96 6 Multi-modal Interventions 97 6.1 INTRODUCTION ……………………………………………………… 98 6.2 BACKGROUND ……………………………………………………….. 99 6.2.1 Circumplex Model of Affect ………………………………….. 99 6.2.2 Auditory Affective Expressivity ………………………………. 100 6.2.3 Haptic Affective Expressivity ………………………………… 100 6.2.4 Multisensory Affective Response ………………………... 100 6.3 PRIOR WORK ……………………...………………….……………… 101 6.3.1 Audio and Haptic Research …………………………………. 101 6.3.2 Vibrotactile Interfaces ………………………………………… 101 6.4 EXPERIMENT DESIGN ……………………………………………… 102 6.4.1 Vibrotactile Wearable Hardware …………………………….. 102 6.4.2 Stimuli Generation …………………………………………….. 102 6.4.3 Conditions ……………………………………………………… 104 v 6.5 RESULTS ……………………………………………………………… 104 6.5.1 Quantitative Analysis …………………………………………. 104 6.5.2 Qualitative Analysis …………………………………………… 109 6.5.3 Usability Assessment ………………………………………… 111 6.6 IMPLICATIONS FOR DESIGN ……………………………………… 113 6.7 FUTURE WORK ………………………………………………………. 114 6.8 CONCLUSION ………………………………………………………… 115 7 Urban Ambien Interventions – Fiat Lux 116 7.1 INTRODUCTION ……………………………………………………… 117 7.2 PREVIOUS WORK …………………………………………………… 118 7.2.1 Urban Interactive Light Design ………………………………. 118 7.2.2 Light Interaction